Cellulose is the world’s most abundant natural, renewable and biodegradable polymer. Lignocellulosic wastes are pretreated and then subjected to an enzyma<c step to obtain an hydrolysate which can be used as C source by bacterial strains [7]. Biodegradable polyesters from agro&industrial by-products Catarina Dias de Almeida 1,2 , João Cavalheiro 1 , Rodrigo Raposo 1 , Teresa Cesário 1 , Frederico Ferreira 1 , Bruno Ferreira 3 , Frederik van Keulen 3 , Eric Pollet 4 , M. Manuela R. da Fonseca 1 1 InsJtute for Biotechnology and Bioengineering, InsJtuto Superior Técnico, Universidade de Lisboa, Portugal, 2 Centro de InvesJgação Interdisciplinar Egas Moniz, Monte de Caparica, Portugal, 3 Biotrend S.A., Cantanhede, Portugal 4 Ecole Europeènne de Chimie, Polymères et Matériaux, Université de Strasbourg, France [email protected] Introduction Results and Discussion Methodology Conclusions Objective Conversion of waste glycerol and hydrolyzed straw into biodegradable, biocompatible polyesters through up-scalable, high productivity processes References [1] Lee, S. Y. Biotechnol. Bioeng. 49, 25-30, 1996. [2] Steinbüchel, A. In Rehm, H.J., Reed, G., Pühler, A., Stadler, P. (eds), Biotechnol. 6, 403-464, 1996. [3] http://www.ebb-eu.org/biodiesel.php (10-04-2014) [4] Solaiman et al. Appl. Microbiol. Biotechnol. 71, 783 – 789, 2006. [5] Koller et al. Biomacromol. 6, 561‑565, 2005. [6] Mothes et al., Eng. Life Sci. 7, 5, 475-479, 2007. [7] Almeida et al., J. Chem. Tech. Biotechnol., 82, 4, 340–349, 2007. [8] Cavalheiro et al., Process Biochem. 44, 509-515, 2009. [9] Cavalheiro et al., Biores. Technol., 111, 391-397, 2012. [10] Lee et al., J. of Biosci. Bioeng, 89 (4), 380-383, 2000. [11] Cavalheiro et al. Biores. Technol., 147, 434–441, 2013. [12] Canadas et al., Int. J. Biol. Macromol., accepted, 2014. [13] Cesário et al., New Biotechnol., 31, 1, 104-113, 2014. [14] Cesário et al. Int. J. Biol. Macromol., accepted, 2014. Waste glycerol was successfully used as an alterna<ve Csource to produce PHAs. Integra<ng biopolyesters and biodiesel produc<on can contribute to (i) reduce costs associated with the Csource for PHAs produc<on (ii) the upgrade of the surplus glycerol generated in the biodiesel plants. P(3HB) and P(3HB4HB) were successfully bioproduced from different batches of wheat straw hydrolysates. The aOained P(3HB) and P(3HB4HB) volumetric produc<vi<es are by far the highest ever achieved on agricultural waste hydrolysates and further op<miza<on is under way (see poster Cesário et al. for more details!). Polyhydroxyalkanoates (PHAs) are biodegradable, compostable polyesters synthesized by many microbial strains under unbalanced growth condi<ons [1] [2]. Under appropriate feeding strategies, copolymers with a wide range of proper<es can be produced, e.g. for agricultural, biomedical and packaging applica<ons. The most common type of PHA is the homopolymer poly(3hydroxybutyrate) (P(3HB)). Copolymers such as P(3HBco4HB) exhibit higher elas<city, lower crystallinity and present more adequate proper<es for specific uses. Current biodiesel produc<on is based on transesterifica<on of vegetable oils genera<ng FAMEs and a surplus of waste glycerol (GRP) (1 t of GRP for each 9 t of biodiesel). Today, approx. 120 plants produce up to 6 100 thousand tonnes of biodiesel annually in the EU, genera<ng 600 thousand t of GRP [3]. GRP can be used as the major C source for PHA produc<on [4][6]. Cupriavidus necator DSM 545 produced the homopolymer P(3HB) from GRP at a produc<vity of 1.1 gL 1 .h 1 [8]. Using γbutyrolactone (GBL) as the precursor for 4HB monomers, C. necator DSM 545 cells were able to accumulate P(3HBco4HB) with different 4HB to 3HB ra<os [9]. Propionic acid (PA) was used as s<mulator for 4HB incorpora<on, as suggested by Lee et al. 2000 [10]. PA considerably increased the 4HB ra<o, but also acted as 3hydroxyvalerate (3HV) precursor, resul<ng in the produc<on of P(3HB4HB3HV). By manipula<ng the dissolved oxygen concentra<on (DOC) and cul<va<on <me, 4HB molar percentages in the range 11.4 21.5 were aOained. Terpolymers were obtained with 24.8% to 43.6% 4HB and 5.6% to 9.8% 3HV (see Table 1, Figure 1 and 2). Results indicate that a higher DOC favors PHA accumula<on. Average MW varied between 5.5 x 10 5 Da and 1.37 x 10 6 Da with a PI from 2.6 to 4.0. Upgrading by-products Waste glycerol Figure 1. Average composiJon of lignocellulosic biomass and major derived hydrolysis products [7] Figure 3. ProducJon of P(3HB4HB3HV); CulJvaJon Jme course for the producJon of P(3HB4HB) with the DOC controlled at 20%. AccumulaJon phase – 2 gL1 of propionic acid were added at the beginning and GRP and GBL were conJnuously fed. Figure 4. PHA (%) accumulated per total DW, by cells of C. necator DSM 545 fed with GRP and GBL in culJvaJons with the DOC controlled at 2 % or 20 % in the presence or absence of 2 gL1 of PA. 0 5 10 15 20 25 30 35 40 0 10 20 30 40 50 60 70 PHA % Time (h) 2 gL1 PA; DOC 20% 2 gL1 PA; DOC 2% DOC 20% DOC 2% Table 1. PHA producJon effect of DOC (%), PA and culture Jme on PHA producJvity, concentraJon and composiJon [9]. Values at the time of max volumetric productivity Values at the end of the cultivation Accum. conditions PHA ProdVol gL -1 h -1 PHA PHA 4HB 3HV ProdVol gL -1 h -1 PHA PHA 4HB 3HV DOC (%) Supplements type [ time (h) ] gL -1 % % % [ time (h) ] gL -1 % % % 20 GBL P(3HB-4HB) 0.33 9.7 25.9 11.4 - 0.17 10.9 36.1 17.6 - [ 29.3 ] [ 64.3 ] 2 GBL P(3HB-4HB) 0.15 4.0 14.8 12.3 - 0.06 4.2 17.9 21.5 - [ 26.0 ] [ 67.5 ] 20 GBL + PA P(3HB-4HB-3HV) 0.29 10.8 29.5 26.2 9.8 0.25 16.7 36.9 43.6 6.0 [ 38.0 ] [ 67.0 ] 2 GBL + PA P(3HB-4HB-3HV) 0.35 11.0 25.0 30.6 6.7 0.21 9.3 18.1 24.8 5.6 [ 31.8 ] [ 44.0 ] The proper<es of the polymers considerably differ from those of the homopolymer P(3HB), even at low percent incorpora<on of 4HB monomers in the polyester chain [11]. Evidence (obtained by DSC and SEC) demonstrates the biosynthesis of heterogeneous PHA blends composed by various frac<ons of different molar composi<on, instead of a narrow distribu<on of chains with similar composi<on [11]. 1 2 3 4 5 1 3 6 5 PHA accumulation – N-limitation or P-limitation as trigger for PHA accumulation C-source quantification – by HPLC PHA recovery – CHCl 3 extraction followed by precipitation with C 2 H 5 OH PHA and GBL quantification – GC PHA characterization – H 1 -NMR, SEC, DSC Wastes as C-source Selected Bacterial strains 2L STR fermentation on-line data Lyophilized cells Extracted polymer Nile-blue stained PHA granules inside C. necator cells Figure 2. Bench scale methodology Poly(3hydroxybutyrate) (P(3HB)) H HO Poly(3HBco4HB) H HO 0 2 4 6 8 10 12 14 16 18 20 0 5 10 15 20 25 30 35 40 45 50 0 10 20 30 40 50 60 70 PHA, 3HB, 4HB, 3HV (gL 1 ) Dry Weight (gL 1 ) Time (h) DW Res DW PHA 3HB 4HB 3HV accumulaJon phase Figure 6. From P(3HB4HB3HV) to scaffolds for hMSC proliferaJon [12]. Acknowledgements Studies on GRP were financed by the EU Integrated Project BIOPRODUCTION (contract nº 026515-2) and those on cellulosic hydrolysates are funded by the EU Collaborative Project BUGWORKERS (contract nº 246449) C. Almeida, J. Cavalheiro, T. Cesário and F. Ferreira were supported by Fundação para a Ciência e Tecnologia, Portugal (SFRH/BPD/26678/2006, SFRH/BD/45266/2008 and SFRH/BPD/ 68587/2010, IF/00442/2012). Waste glycerol was kindly provided by Torrejana, Fábrica de Biocombustíveis (Torres Vedras). Lignocellulosic hydrolysates were produced by biorefinery.de GmbH, Germany, in the framework of BUGWORKERS project. PRODUCTION bio Figure 5. The effect of (4HB + 3HV) molar % on the terpolymer mechanical properJes: (a) and (b) elongaJon at break ploled on two different scales, (c) Young’s modulus and (d) tensile strength. Polyhydroxyalkanoates Wheat straw lignocellulosic hydrolysates A standardized fermenta<on protocol was successfully developed allowing for the tes<ng of the supplied hydrolysates, as well as for further scale up of the process. Feedback from bench scale assays allowed for hydrolysates improvement by biorefinery.de GmbH. Feeding strategy for mul<ple carbon source consump<on successfully implemented [13]. Cell density and P(3HB) produc<vity were similar to those reached in control cul<va<ons with mixtures of commercial sugars [13]. Cell density and P(3HB4HB) produc<vity using hydrolysates were lower than those reached in P(3HB) produc<on cul<va<ons due to inhibi<on caused by the precursor (GBL) [14]. Table 2. Preliminary strain selecJon. Table 4. PHA producJon from wheat straw hydrolysates (hydrolysate composiJon: 465 gL1 glucose, 146 gL1 xylose, 42 gL1 arabinose). Biomass (milled wheat straw) Concentrated hydrolysate AFEX ENZYMATIC HYDROLYSIS WATER EVAPORATION Csource Y P (3HB)/S Prod vol (gL 1. h 1 ) P (3HB) (%) Dw (gL 1 ) t (h) Sugar mix 0.18 1.7 51 138 36.5 Hydrolysate 0.20 1.7 55 150 38.5 Table 3. P(3HB) producJon from a sugar mixture simulaJng the hydrolysate (as control) and from a real hydrolysate (hydrolysate composiJon: 563 gL1 glucose, 284 gL1 xylose, 46 gL1 arabinose). Polymer CDW (gL -1 ) PHA (gL -1 ) PHA (%) Prodvol (gL -1. h -1 ) P(3HB) 40h 125 71 57 1.5 P (3HBco 6% 4HB) 49h 88 24 27 0.5 Figure 7. Main steps involved in hydrolysate producJon at biorefinery.de GmbH Strain Criteria Comments glucose xylose PHA Risk Alcaligenes latus DSM 1122 + - + 1 Aggregates during the first stages of growth Bacillus sp MA3.3 + + + 1 Not available Burkholderia cepacia ATCC 17759 + + + 2 Burkholderia sacchari DSM 17165 + + + 1 High growth rates Cupriavidus necator DSM 545 + - + 1 High growth rates Haloferax mediterranei + - + 1 Costly growth medium and corrosion problems. Methylobacterium extorquens + - + 1 Low concentration of the optimal C sources in the hydrolysates Sphingopyxis macrogoltabida LMG 17324 + - + 1 Preferential uptake of oligosaccharides